Hacker News new | ask | show | jobs
by bmj1 4482 days ago
Background: I spent the last 2.5 years working as a product manager at another payments company. I've integrated Stripe.js in various side projects. I saw first hand what happens when you make a checkout highly customisable (dozens of options): you introduce huge complexity cost.

Here are some ways this manifests:

- longer dev cycles/increased difficulty in testing/more frequent release rollbacks

- higher merchant support costs

- harder to innovate

- a need to support dozens of legacy use-cases that only some merchants use

- harder to onboard new merchants

- more complex sales process

- inconsistent user experience for customers

- higher staff training costs/time

- impossible to optimise conversion for the majority, as each merchant is showing a different checkout.

It all starts with 1 option, but it's a slippery slope.

In summary, I think Stripe are probably taking the right approach here, and I wish them the best of luck.